Background: Stress-only imaging saves time and radiation exposure, but apprehension remains about the reliability, diagnostic, and prognostic accuracy of a normal stress-only study. The objective of this study was to determine the prognosis of stress-only SPECT MPI in routine clinical practice.
Methods: Patients at lower pre-test risk for CAD presenting for a Tc-99m SPECT MPI over a 2-year period underwent a stress-only protocol. If the stress images were normal (attenuation correction was routinely acquired on all patients), rest imaging was not done. Outcomes of the stress-only group were compared to a full rest-stress protocol cohort. Only patients with normal perfusion and left ventricular function, and no known CAD, were included. All-cause mortality was determined using the Social Security Death Index and specific causes of death were determined using the National Death Index. The difference in all-cause and cardiac mortality between groups in the presence of competing risks was assessed using log-normal survival models.
Results: Out of 10,609 patients studied during the time period, 1,673 had a normal stress-only study and 3,237 had a normal rest-stress study. At one year, there were 20 total and 3 cardiac deaths (1.2% and 0.2% mortality) in the stress-only group, and 40 total and 4 cardiac deaths (1.2% and 0.1% mortality) in the rest-stress cohort. At the end of follow-up (40 +/- 9 months), there were 46 total and 7 cardiac deaths (2.7% and 0.4% mortality) in the stress-only group, and 119 total and 17 cardiac deaths (3.7% and 0.5% mortality) in the rest-stress cohort. No significant difference between the stress-only and rest-stress cohorts was found after controlling for confounding variables for both all-cause mortality (p = .94) and cardiac mortality (p = .82).
Conclusions: A normal stress-only MPI has an excellent short-term prognosis (both for all-cause and cardiac mortality) comparable to that of a normal rest-stress MPI study.